Age and albumin D site-binding protein control tissue plasminogen activator levels: neurotoxic impact

Abstract

Recombinant tissue-type plasminogen activator (tPA) is the fibrinolytic drug of choice to treat stroke patients. However, a growing body of evidence indicates that besides its beneficial thrombolytic role, tPA can also have a deleterious effect on the ischaemic brain. Although ageing influences stroke incidence, complications and outcome, age-dependent relationships between endogenous tPA and stroke injuries have not been investigated yet. Here, we report that ageing is associated with a selective lowering of brain tPA expression in the murine brain. Moreover, our results show that albumin D site-binding protein (DBP) as a key age-associated regulator of the neuronal transcription of tPA. Additionally, inhibition of DBP-mediated tPA expression confers in vitro neuroprotection. Accordingly, reduced levels of tPA in old mice are associated with smaller excitotoxic/ischaemic injuries and protection of the permeability of the neurovascular unit during cerebral ischaemia. Likewise, we provide neuroradiological evidence indicating the existence of an inverse relationship between age and the volume of the ischaemic lesion in patients with acute ischaemic stroke. Together, these results indicate that the relationship among DBP, tPA and ageing play an important role in the outcome of cerebral ischaemia.

Figure 1

Effect of ageing on tPA proteolytic activity and transcription in the CNS and the plasma. tPA proteolytic activity was normalized to mean values of 1-month-old C57BL6/J female mice in the perfused cortex (A), hippocampus (B) and spinal cord (C) (n = 4 per group). (D) Similar experiments performed on plasma samples from mice aged 2 and 18 months (n = 6 per group). Levels of tPA transcription analysed by Q-RT–PCR were normalized to mean value of 1 month old C57BL6/J female mice in the perfused cortex (E) and hippocampus (F) (n = 4 per group). *Significantly different from 1-month-old mice, P < 0.05.

Figure 2

Effect of ageing on tPA protein levels in the CNS. Control immunostainings for tPA (red) in WT and tPA-deficient mice (A; n = 3). Immunostainings for tPA (red) on perfused brain sections of 4-, 12- and 21-month-old WT mice (B) and corresponding nuclei counting using DAPI (blue) counterstaining (C; n = 3 per group). Scale bars, 100 µm. ELISA for total tPA performed on perfused brain extracts of WT mice at 4- and 21-months-old (cortex and hippocampus) (D; n = 4). *P < 0.05.

Figure 3

Effect of ageing on DBP's transcription and protein levels in the cortex. tPA promoters analysis for the transcription factor DBP (albumin D-site binding protein) sequence in the different species analysed (A). Q-RT–PCR analysis for DBP in the perfused cortex (B) or hippocampus (C) as a function of age (n = 4). (D) The positive correlation between DBP and tPA mRNA cortical levels during ageing (blue squares: young animals; red squares: old animals). Immunostaining for tPA and DBP (E) in the perfused hippocampus with positive staining in mossy fibres and neuronal bodies, respectively. *Significantly different from 3-month-old mice, P < 0.05.

Figure 4

Effect of ageing on ischaemia-induced blood–brain barrier leakage and infarct volumes. (A) Mean volume of the ischaemic lesion 48 h after MCAO in WT mice 2–24 months of age (n = 6 for each observation). *P < 0.05. (B) Evans blue dye extravasation 48 h after MCAO in WT mice 2–24 months of age (n = 6 for each observation). *P < 0.05. (C) Q-RT–PCR analysis for tPA, NR1 and β-actin 48 h after MCAO in WT mice at 4 and 21 months of age (n = 4 for each observation; histograms represent the ratio between mRNA levels in the ischaemic versus non-ischaemic cortex). (D) Illustrative photomicrographs of the ischaemic lesion in WT mice at 4 and 21 months of age [upper panel: thionin staining; lower panel: immunohistochemistry for MAP-2 (green) and GFAP (red); Scale bar: 500 µm]. (E) Mean volume of the ischaemic lesion 48 h after MCAO in WT, WT mice injected with rtPA, tPA^–/– mice, tPA^–/– mice injected with rtPA at 4 and 21 months of age (n = 6 for each observation). *P < 0.05 versus WT and tPA^–/–treated mice, ^†P < 0.05 versus tPA^–/– mice (4 months old), ^‡P < 0.05 versus WT and untreated tPA^–/– mice (21 months old). (F) Evans blue extravasation 48 h after MCAO in WT or tPA^–/– mice (4 and 21 months of age), injected or not with rtPA (n = 6 for each observation). *P < 0.05 versus tPA^–/– mice (4 months old), *P < 0.05 versus tPA^–/– and WT mice (4 and 21 months old). M: age in months.

Figure 5

Age-related reduction in the volume of the ischaemic lesion in patients with acute ischaemic stroke and effect of thrombolysis. (A) Distribution of ages among the studied stroke population. (B) Proportion of patients having a lesion volume larger than 14cc at DWI performed at arrival regarding age decades. (C) Correlation between neurological scores (NIHSS) and infarct extension (DWI) both measured at stroke patient arrival at the emergency department. (D) MRI of two representative cases of human MCAO with ischaemic infarct observed at emergency department arrival (within 3 h of stroke onset). Both cases have similar location of the MCAO, however, upper case (36 years) has a measured DWI volume of 21cc as compared with bottom case (79 years) with DWI volume of 12cc at arrival.

Figure 6

Effect of DBP repression on tPA transcription and proteolytic activity in neuronal cultures. Cultured cortical nurons were transiently transfected with either double stranded DBP-decoy or scramble oligonucleotides (A–C) or shRNAs targeting the expression of DBP (D–F) (n = 18 from three independent cultures per group). DBP decoy reduced tPA mRNA levels measured by Q-RT–PCR (A) and active tPA levels in the corresponding bathing media assessed by zymography (B) (n = 18 from three independent cultures per group). DBP decoy-transfected neurons were less sensitive to NMDA-mediated excitotoxicity than DBP scramble-transfected neurons (C). shRNAs reduced the transcription of DBP (D) and tPA (E), while the empty vector (Ctrl Plok) had no effect. Neuronal death induced by NMDA exposure was prevented by DBP shRNAs (F) (n = 12 from three independent cultures per group). *P < 0.05 **P < 0.01, ***P < 0.0001.